We present a small study analyzing how prompt token classification loss weighting (PLW) affects the performance of 7B-size LLaMA models fine-tuned on instruction tasks. We recreated Stanford's Alpaca experiment with both LLaMA 1 and LLaMA 2 using multiple instruction datasets. We found that models fine-tuned on our short-completion dataset have a negative quadratic relationship with PLW while models fine-tuned on long-completion datasets were unaffected by PLW.

翻译：我们开展了一项小型研究，分析提示词令牌分类损失加权（PLW）如何影响在指令任务上微调的7B规模LLaMA模型性能。我们使用多个指令数据集重现了斯坦福大学的Alpaca实验（同时采用LLaMA 1和LLaMA 2）。研究发现，基于短补全数据集微调的模型与PLW呈现负二次关系，而基于长补全数据集微调的模型则不受PLW影响。